CropImageReconition_TL
This project demonstrates an approach of transfer learning which can achieve crop images recognition. In our case, we employed three types of pre-trained model. They are ResNet50, VGG16 and MobileNet respectively. The objective of this project is to find out the most suitable pre-trained model for this scenario, judging from their performance on testing set.
Objective
The dataset includes four types of crop image which are Rice, Banana tree, Green onion and Sugarcane, and we anticipated to achieve image recognition using Deep Learning. Besides, we decided to employ a common strategy known as Transfer Learning.
Implementation
Source code are stored in script folder, and experiment record is stored on log folder.